1. Field of the Invention
The present invention relates to a semiconductor wafer with improved flatness in the peripheral area and to a process for producing a semiconductor wafer of this nature. Semiconductor wafers with a high degree of flatness are suitable for use in the semiconductor industry, in particular for the manufacture of electronic components with line widths of less than or equal to 0.13 xcexcm
2. The Prior Art
A semiconductor wafer which is intended to be suitable in particular for the manufacture of electronic components with line widths of less than or equal to 0.13 xcexcm must have a large number of special properties. A particularly important property is the local flatness of the semiconductor wafer. The modern stepper technology requires optimum local flatness in all partial areas of a side of the semiconductor wafer. The back-surface-referenced global (GBIR; previously also known as TTV) and local (SBIR; previously also known as LTV) flatness measurements which were in widespread use in the past are being replaced by flatness measurements which take into account the focusing ability of a stepper in all partial areas of the wafer surface. Such a flatness measurement is the SFQR (site front-surface referenced least squares/range=range of the positive and negative deviation from a front surface defined by minimizing the mean square error for a component area of defined size). The value SFQRmax indicates the maximum SFQR value for all the component areas on a semiconductor wafer. A generally accepted rule of thumb states that the SFQRmax value of a semiconductor wafer must be less than or equal to the possible line width on this wafer for semiconductor components which can be produced on it. If this value is exceeded, the stepper experiences focusing problems and the component in question will be lost.
The final flatness of a semiconductor wafer is generally produced by a polishing process. Equipment and processes for the simultaneous polishing of front and back surfaces of a semiconductor wafer, in order to improve the flatness values of the semiconductor wafer, have been put in place and refined. This so-called double-side polishing is described, for example, in U.S. Pat. No. 3,691,694.
Double-side polishing is also described in EP 208 315 B1. Semiconductor wafers are in carriers which are made from metal or a plastics material and have suitably dimensioned cutouts. The wafers and carriers are moved along a path which is predetermined by the machine and process parameters between two rotating polishing plates. These plates are covered with a polishing cloth, and are in the presence of a polishing fluid; and the wafers are thusly polished (in the specialist literature, carriers are also referred to as templates). The pressure forces which are employed during double-side polishing act preferentially on the semiconductor wafer which is to be polished and not on the carrier. To achieve this result, the end thickness of semiconductor wafers which have undergone double-side polishing according to the prior art is considerably thicker than that of the carriers used. For example, E. Mendel and J. R. Hause, in IBM Technical Report TR22.2342, presented at the Spring Meeting of the Electrochemical Society in Boston, Mass. on 05.10.1979, recommend that the wafers project by 2 to 3 mil (corresponds to 51 to 76 xcexcm). This degree of projection can be ensured either by establishing the polishing time required on the basis of the material-abrasion rates determined in preliminary tests for a specific test procedure or by means of spacers arranged on the carrier. This is described in U.S. Pat. No. 5,422,316.
It is known to integrate double-side polishing into process sequences for producing semiconductor wafers. EP 754 785 A1 describes the sequence of sawing a semiconductor crystal, followed by edge rounding, double-side polishing and final polishing of the semiconductor wafers obtained. In EP 755 751 A1, it is proposed to employ a double-side grinding process between the edge rounding and the double-side polishing. The preferred embodiments described in U.S. Pat. No. 5,756,399 include the process sequence of sawingxe2x80x94edge roundingxe2x80x94grindingxe2x80x94alkaline etchingxe2x80x94double-side polishing.
DE198 33 257 C1 describes the process sequence of sawingxe2x80x94edge roundingxe2x80x94grindingxe2x80x94etchingxe2x80x94double-side polishingxe2x80x94final polishing. The etching in this case is carried out using an improved acid etching process. A common feature of these process sequences is that after the double-side polishing they lead to a semiconductor wafer which has higher local geometry values, expressed as SFQR, in the peripheral area than in the central area of the semiconductor wafer. Under certain conditions procedures such as that described in EP 187 307 A1, are aimed at compensating for the thinner edges by providing an initial dish-like geometry. This may lead to an improvement in the rear-surface-referenced geometry values GBIR and SBIR. However, such procedures represent a disadvantage with regard to the SFQR values which are relevant for the production of components.
It is an object of the present invention to provide a semiconductor wafer which is suitable in particular for the manufacture of electronic components with line widths of less than or equal to 0.13 xcexcm and which does not have the abovementioned drawbacks with regard to the local geometry in the peripheral area, expressed as SFQR. Furthermore, the further properties of the semiconductor wafer should be at least equal to those of semiconductor wafers produced according to the prior art, and there should be advantages in terms of production costs.
The present invention is directed to a semiconductor wafer with a front surface and a back surface and flatness values based on partial areas of a surface grid on the front surface of the semiconductor wafer, which has a maximum local flatness value SFQRmax of less than or equal to 0.13 xcexcm and individual SFQR values which in a peripheral area of the semiconductor wafer do not differ significantly from those in a central area of the semiconductor wafer.
Preferably, in this context, there is considered to be no significant difference between individual SFQR values in the peripheral area and the central area if the arithmetic mean of the SFQR values for the peripheral area differs by at most 0.03 xcexcm from the arithmetic mean of the SFQR values for the central area.
The present invention is also directed a process for producing a semiconductor wafer by simultaneously polishing of a front surface and a back surface of the semiconductor wafer between rotating polishing plates while a polishing fluid is being supplied, the semiconductor wafer lying in a cutout in a carrier and being held on a specific geometric path, and the carrier being of defined thickness, and the semiconductor wafer having a starting thickness prior to polishing and an end thickness after polishing, wherein the starting thickness of the semiconductor wafer is 20 to 200 xcexcm greater than the thickness of the carrier and the semiconductor wafer is polished until the end thickness of the semiconductor wafer is 2 to 20 xcexcm greater than the thickness of the carrier.
The essential feature of the invention is that the semiconductor wafer must be subjected to double-side polishing until the thickness of the final polished semiconductor wafer is only slightly greater than the thickness of the carrier used. The thickness difference must lie within a narrow window of from 2 to 20 xcexcm. As the examples given below will illustrate, the discovery of this window was surprising and impossible to foresee.
The starting material used for the process is a semiconductor wafer which has been separated in a known way from a crystal. For example it may be separated from a silicon single crystal which has been cut to length and undergone circular grinding and the front and/or back surfaces of which have undergone a surface-grinding step. If desired, the crystal may be provided with one or more orientation features in order to be able to identify the crystal axes, for example a notch and/or a flat. The edge of the semiconductor wafer may also be rounded at a suitable point in the process sequence by means of a suitably profiled grinding wheel. Moreover, it is possible for the surface of the semiconductor wafers to be etched after the grinding step.
The end product of the process is a semiconductor wafer which has undergone double-side polishing and satisfies the requirements imposed on semiconductor wafers to be used as starting material for semiconductor component processes with line widths of less than or equal to 0.13 xcexcm. And due to the reduced abrasion of material, the wafer has advantages in terms of lower production costs over the semiconductor wafers which are produced according to the prior art.
In principle, the process according to the invention can be used to produce a wafer-like body which comprises a material which can be machined with the chemo-mechanical double-side polishing process used. Examples of such materials, which are processed further predominantly in the semiconductor industry, without being limited to this industry, are silicon, silicon/germanium, silicon dioxide, silicon nitride, gallium arsenide and further so-called III-V semiconductors. Silicon in single-crystal form, for example crystallized by a Czochralski growing process or a float zone growing process, is preferred. Silicon with a crystal orientation (100), (110) or (111) is particularly preferred.
The process is particularly suitable for the production of silicon wafers with diameters of, in particular, 200 mm, 300 mm, 400 mm and 450 mm and thicknesses from a few 100 xcexcm to a few cm, preferably from 400 xcexcm to 1200 xcexcm. The semiconductor wafers can either be used directly as a starting material for the production of semiconductor components or can be supplied for their intended purpose after a final polishing step according to the prior art has been carried out. Also, the wafers can be supplied after the application of layers such as back surface seals or an epitaxial coating on the wafer front surface, using silicon or other suitable semiconductor materials. Also, the wafer can be supplied after conditioning by means of a heat treatment, for example under a hydrogen or argon atmosphere. Of course, the invention can be applied not only to the production of wafers composed of a homogeneous material but also to the production of layered semiconductor substrates. These substrates are of multilayer structure, such as SOI (silicon-on-insulator) wafers and so-called bonded wafers.
The process is described further with reference to the production of a silicon wafer.
It is possible for a silicon wafer which has been sawn, for example, by means of an annular sawing or wire sawing process and which, depending on the diameter and the nature of the sawing process, exhibits damage down to a depth in the range from 10 to 40 xcexcm, to be subjected directly to the double-side polishing step according to the invention. However, it is preferred for the sharp, and therefore highly mechanically sensitive edge of the wafer to be rounded with the aid of a suitably profiled grinding wheel. Furthermore, it is possible, in order to improve the geometry and partially abrade the destroyed crystal layers, to subject the silicon wafer to a mechanical abrasion step, such as lapping or grinding, in order to reduce the amount of material removed in the polishing step according to the invention. It is preferable for the silicon wafer to be subjected to a surface-grinding step, in which either one side is ground or both sides are ground sequentially or both sides are ground simultaneously. Sequential surface grinding of the front and back surfaces of the wafer is particularly preferred. To remove the damage to the wafer surface and edge which has inevitably been caused by the mechanical process steps and to remove any impurities which may be present, for example metal impurities which are bonded in the damaged area, an etching step may take place at this point. This etching step may be carried out either as a wet-chemical treatment of the silicon wafer in an alkaline etching mixture or acidic etching mixture or as a plasma treatment. An acid etching step in a mixture of concentrated aqueous nitric acid and concentrated aqueous hydrofluoric acid according to the disclosure of DE 198 33 257 C1 is preferred.
A particularly preferred starting material for the polishing process according to the invention is semiconductor wafers which are made from silicon and have a diameter of greater than or equal to 200 mm. These wafers are produced by sawing a single silicon crystal, followed by edge rounding, sequential surface grinding of both sides of the wafer, and removing from 10 xcexcm to 100 xcexcm of silicon per side. This is followed by wet-chemical etching in an acidic etching mixture, removing from 5 xcexcm to 50 xcexcm of silicon per side of the wafer.
A commercially available double-side polishing machine of suitable size, as described, for example, in IBM Technical Report TR22.2342, can be used to carry out the polishing step according to the invention. The polishing machine essentially comprises a bottom polishing plate which can rotate freely in the horizontal plane and a top polishing plate which can rotate freely in the horizontal plane. Both of these plates are covered with a polishing cloth, and, given a continuous supply of a polishing fluid of suitable chemical composition, this machine allows abrasive polishing of semiconductor wafers, in this case silicon wafers, on both sides.
It is possible to polish only one silicon wafer. For cost reasons, however, it is usual to polish a large number of silicon wafers at the same time, the number depending on the design features of the polishing machine. In this case, the silicon wafers are held on a geometric path which is determined by machine and process parameters, and by means of carriers which have suitably dimensioned cutouts for holding the silicon wafers. The carriers are in contact, for example by way of a pin gearing or an involute gearing, with the polishing machine, via a rotating inner pin ring or gear ring and an outer pin ring or gear ring which generally rotates in the opposite direction. Thus they are caused to rotate between the two polishing plates.
Examples of parameters which influence the path of the silicon wafers in relation to the top and bottom polishing plates during the polishing operation are the dimensions of the polishing plates, the design of the carriers and the rotational speeds of the top polishing plate, the bottom polishing plate and the carrier. If there is one silicon wafer in the center of a carrier, the silicon wafer moves in a circle around the center of the polishing machine. If a plurality of silicon wafers are arranged eccentrically in a carrier, the rotation of the carriers about their own axis results in a hypocycloidal path. A hypocycloidal path is preferred for the polishing process according to the invention. It is particularly preferable for from four to six carriers, which each hold at least three silicon wafers arranged at regular intervals on a circular path, to be used simultaneously.
The carriers used in the process according to the invention may be made from any material which exhibits sufficient mechanical stability when presented with the mechanical stresses, primarily the compressive and tensile stresses, caused by the drive. Moreover, the material must not be chemically and mechanically attacked to any significant extent by the polishing cloths and the polishing fluid used. This is in order to ensure a sufficiently long service life for the carriers and to avoid contamination of the polished silicon wafers. Furthermore, the material must be suitable for the production of highly planar carriers which are free from stresses and corrugations and are of the desired thickness and geometry. In principle, the carriers may, for example, be made from metal, a plastics material, fiber-reinforced plastic or plastic-coated metal. Carriers made from steel or from fiber-reinforced plastic are preferred; carriers made from stainless chrome steel are particularly preferred.
The carriers have one or more cutouts, preferably in the shape of a circle, for holding one or more silicon wafers. To ensure that the silicon wafer can move freely in the rotating carrier, the diameter of the cutout must be slightly greater than that of the silicon wafers which are to be polished. A diameter which is from 0.1 mm to 2 mm larger is preferred; a diameter which is from 0.3 to 1.3 mm larger is particularly preferred. To prevent damage to the edge of the wafer from the inner edge of the cutout in the carrier during polishing, it is sensible, and therefore preferred, for the inner side of the cutouts to be lined with a plastic coating of the same thickness as the carrier. This is described in EP 208 315 B1. Examples of suitable plastics materials are polyamide, polyethylene, polypropylene, polyvinyl chloride, polytetrafluoroethylene or polyvinylidene difluoride, which are all equally preferred. However, it is also possible, in accordance with the procedure described in EP 776 030 A2, to provide the inner side of the cutouts with a device which allows simultaneous polishing of the edge of the silicon wafer during the double-side polishing.
The preferred thickness of the carriers for the polishing process according to the invention is from 400 to 1200 xcexcm, depending on the final thickness of the polished silicon wafers, which is ultimately dependent on the diameter of the silicon wafers and on the planned application. The characteristic feature of the invention is that the starting thickness of the silicon wafers which are to be polished is preferably 20 to 200 xcexcm greater than the thickness of the carrier, with the range from 30 to 70 xcexcm being particularly preferred. Also, the end thickness of the polished wafers is preferably from 2 to 20 xcexcm greater than the thickness of the carrier, with the range from 5 to 15 xcexcm being particularly preferred. However, slight deviations from the thickness ranges indicated are possible without any serious increase in the local flatness values SFQR being observed in the peripheral area of the wafers. The amount of silicon removed by the polishing step is from 5 to 100 xcexcm, preferably from 10 xcexcm to 60 xcexcm, and particularly preferably from 20 to 50 xcexcm in thickness.
In view of the statements made with regard to the thickness ratios, the double-side polishing step is preferably carried out in the way which is known to the person skilled in the art. Polishing cloths with a wide range of properties are commercially available. Polishing is preferably carried out using a commercially available polyurethane polishing cloth with a hardness of from 40 to 120 (Shore A). Polyurethane cloths with incorporated polyethylene fibers, lying within a hardness range from 60 to 90 (Shore A), are particularly preferred. When polishing silicon wafers, it is recommended for a polishing fluid with a pH of preferably 9 to 12, particularly preferably 10 to 11, and preferably comprising from 1% to 10% by weight, particularly preferably 1% to 5% by weight, SiO2 solids content in water, to be supplied continuously. The polishing pressure preferably is from 0.05 to 0.5 bar, particularly preferably from 0.1 to 0.3 bar. The silicon abrasion rate is preferably from 0.1 to 1.5 xcexcm/min, particularly preferably from 0.4 to 0.9 xcexcm/min.
The percent by weight of SiO2 solids content is based upon the total polishing fluid weight.
After polishing is complete, residual polishing fluid is cleaned off the silicon wafers if necessary. Then the wafers are dried and their local geometry SFQR can then be measured on a commercially available geometry gauge which operates, for example, capacitively or optically.
Depending on their intended application, it may be necessary to subject the front surface of the wafers to final polishing according to the prior art. For example, a soft polishing cloth may be used with the aid of an alkaline polishing fluid based on SiO2. To obtain the very low, uniformly distributed local geometry values produced in the polishing step according to the invention, the amount of silicon removed from the wafer in this final polishing should be relatively low, for example around 0.1 to 0.7 xcexcm in thickness.
If necessary, the semiconductor wafer may be subjected to a heat treatment at a suitable point in the process sequence, for example in order to destroy thermal donors or to anneal flaws in crystal layers which are close to the surface. Furthermore, a laser-marking step in order to identify the wafer and/or an edge-polishing step may be added at a suitable point in the process sequence. For example this may occur before or after grinding in the case of laser marking and before, during or after double-side polishing in the case of edge polishing. A series of further process steps which are required for certain products, for example the application of surface coatings of polysilicon, silicon dioxide or silicon nitride may be carried out. Also, the application of an epitaxial layer of silicon or further semiconductive materials to the front surface of the silicon wafer can also be incorporated into the process sequence at suitable points. This can be done by using methods which are also known to the person skilled in the art. In addition, it may also be desirable for the semiconductor wafer to be subjected to batch cleaning or individual-wafer cleaning according to the prior art before or after individual process steps.
There are further parameters which are customarily used to characterize wafers and are well known to the person skilled in the art. These parameters include surface defects, roughness and metal contamination of the surface of the wafers, as well as magic-mirror defects. A semiconductor wafer produced according to the present invention is superior to a semiconductor wafer produced according to the prior art.
A semiconductor wafer, in particular a silicon wafer, produced according to the invention fulfills the requirements for the production of semiconductor components with line widths of less than or equal to 0.13 xcexcm. The process according to the invention has proven to be an optimum solution for the production of silicon wafers having the features outlined above. It is surprising and unexpected that only a relatively narrow window for the extent to which the thickness of the finished, polished wafer projects beyond the thickness of the carrier leads to certain unexpected results. These are the elimination of the thin edges, with the associated high SFQR values, which are characteristic of semiconductor wafers which have been subjected to double-side polishing according to the prior art. The geometry conditions imposed on the starting material are minimal, reducing the demands imposed on the preliminary processes. The good geometry which is produced using the step according to the invention is produced even after relatively little material has been removed. Also an improved process reliability, together with a reduced risk of fracture, results in very high yields. This is without cost-intensive steps being required for local correction of the geometry, for example by plasma etching. Therefore, the process of the invention is also economically competitive and can even be used for reworking rejected wafers and to process reclaimed wafers.